Most people are using AI like a smarter version of Google search.
And honestly? That’s fine. It works. But if you’re stopping there, you’re leaving a massive amount of value on the table.
The real unlock isn’t about “using ChatGPT more” or learning a thousand complex prompts. It’s about creating a Custom GPT that actually understands you, your work, and your unique data.
If you’ve ever felt like AI gives you generic, slightly “marketing-bro” sounding answers, this guide is for you. Here is the framework I use to move from generic chats to high-leverage workflows.
The “New Hire” Mindset
Here’s how I think about it:
Imagine you just hired a brilliant new teammate. They’re smart and capable, but they have zero context about your business.
- They don’t know your brand voice.
- They don’t know your audience.
- They don’t know that project you did last year that failed.
If you asked them to “write a strategy,” they’d give you something generic. You wouldn’t expect them to succeed without onboarding, right?
Don’t expect AI to succeed without context, either.
A custom GPT works best when it has three specific things: Context, Data, and Clear Jobs.
1. Context: The “Onboarding” Phase
This isn’t just about typing a prompt; it’s about providing real background. When I build a GPT, I front-load it with the same info I’d give a human hire:
- Your Role: Who are you acting as?
- Your Goals: What are we trying to achieve?
- Your Audience: Who are we talking to? (e.g., “Richmond small business owners,” not “generic customers”).
- Your Constraints: What should we avoid?
- Your Voice: This is critical. Explicitly tell it: “Do not sound like a marketing hype-man. Be conversational, practical, and direct.”
2. Actual Data: Where Efficiency Compounds
This is the step most people skip. A Custom GPT becomes useful when it stops guessing and starts remembering how you think.
Upload your actual files to the knowledge base:
- Docs you reference constantly: PDF guides, brand books, or SOPs.
- Frameworks you reuse: That strategy deck you built last quarter? Upload it.
- Past emails or proposals: Let it see how you actually communicate with clients.
- Notes you’ve never revisited: We all have them. Let the AI parse them so you don’t have to.
3. Clear Jobs to Do
Generic AI answers generic questions. Focused AI does real work.
Instead of asking open-ended questions, I assign my GPTs specific “Jobs to be Done.”
- ❌ Generic: “Write a summary.”
- ✅ Specific: “Turn these messy notes into a client-ready summary.”
- ✅ Specific: “Pressure-test this strategy before I send it. Tell me what I’m missing.”
- ✅ Specific: “Rewrite this in my voice—remove the jargon.”
When you define the job, the output gets dramatically better.
Why This Matters for Leaders
The biggest time drain for most leaders isn’t a lack of ideas. It’s the friction around the ideas.
It’s the time spent:
- Re-explaining context to freelancers or team members.
- Re-finding information you know you saved somewhere.
- Re-deciding things you’ve already decided.
A well-built Custom GPT reduces all three. It doesn’t replace the thinking—it removes the friction around the thinking.
Your Turn
If you’re curious how this could look for your role, your team, or your business, or want to explore the value a custom GPT could unlock for your business, please reach out!
What is the one repeatable, annoying task you’d love to hand off to a Custom GPT first?











